1. Field of the Invention
The present invention relates to carboxy-terminally and carboxy/amino-terminally phosphorylated polyamide nucleic acid (PNA) derivatives having improved properties, to their use and to agents and processes for preparing them.
2. Summary of the Related Art
Polyamide nucleic acids, also termed peptide nucleic acids (PNA), bind to complementary target sequences (DNA or RNA) with a higher affinity than do natural oligonucleotides and, furthermore, have the advantage, as compared with natural DNA, that they are very stable in serum. PNA were originally described as unnatural nucleic acid analogs in which the entire sugar-phosphate backbone is replaced with N-(2-aminoethyl)glycine units (M. Egholm et al. (1991) Science 254, 1497-1500; WO 92/20702; M. Egholm et al. Nature (1993) 365, 566-568; P. Nielsen, (1994) Bioconjugate Chem. 5, 3-7; E. Uhlmann et al. (1998) Angewandte Chemie Int. Ed. Engl. 37, 2796-2823). The bases employed are 1) nucleobases which occur naturally and are customary in nucleotide chemistry, 2) nucleobases which do not occur naturally, and 3) the prodrug forms of these two types of bases, that is, precursors which are only converted into the free base by biotransformation in the body.
PNAs have also been described in which not all the positions in the backbone carry base residues (Greiner et al. (1999) Helv. Chim Acta 82, 2151), and in which aminoethylglycine is replaced by more complex units (Uhlmann et al. (1998) Angewandte Chem. Int. Ed. 37, 2796; Falkiewicz (1999) Biochim. Pol., 46, 509-529).
The fact that the PNA backbone does not have any net charge is a feature of this class of substances that has far-reaching consequences. The fact that PNA binds to complementary DNA and RNA even at low salt concentration (see e.g. Peptide Nucleic Acids: Protocols and Applications; Peter E. Nielsen and Michael Egholm (Edit.) Horizon Scientific Press, 1999, page 3), with the Watson-Crick base pairing rules being obeyed, is ascribed to the neutral character of the PNA and the decrease in charge repulsion which is associated therewith. For this reason, PNA can, in principle, be used for numerous applications in which natural oligonucleotides or oligonucleotide derivatives would otherwise be employed. However, in addition to this, because of the unique binding properties, a large number of applications which are not possible with natural oligonucleotides also ensue (see, for example: Peptide Nucleic Acids: Protocols and Applications; Peter E. Nielsen and Michael Egholm (Edit.) Horizon Scientific Press, 1999). For example, a strand invasion of double-stranded DNA has been observed when using PNA, resulting in formation of triplex structures.
Typical examples of applications for PNA include its use for inhibiting gene expression by binding, in a sequence-specific manner, to cellular DNA or RNA. “Antisense agents” are short, single-stranded nucleic acid derivatives which bind, by means of Watson-Crick base pairing, to a complementary mRNA whose translation into the corresponding protein is to be inhibited (Uhlmann and Peyman (1990) Chem. Rev. 90, 543; Larsen et al. (1999) Biochem. Biophys. Acta 1489,159). “Anti-gene agents” bind, byway of Hoogsteen base pairing, in the major groove of the DNA double helix with the formation of a triple helix, resulting in transcription of the genes being inhibited in a sequence-specific manner (Praseuth et al. (1999) Biochem. Biophys. Acta 1489, 181). Gene expression can also be specifically inhibited by so-called decoy oligomers, which mimic the regions for binding transcription factors. By treating with decoy agents, particular transcription factors can be captured in a sequence-specific manner and activation of transcription thereby prevented (Mischiati et al. (1999) J. Biol. Chem. 274, 33114). Another group of oligonucleotide derivatives which act intracellularly are the chimeraplasts. These are used for specific gene proof-reading (Cole-Strauss et al. (1996) Science 273, 1386-1389).
PNAs can, therefore, be used as pharmaceuticals and/or diagnostic agents or for producing pharmaceuticals and/or diagnostic agents. For example, after having been labeled with biotin, fluorescein, or other labels, PNA can be used as a specific hybridization probe for diagnostic purposes and in molecular biology.
Four methods have so far been described in the literature for introducing the labeling groups (Oerum et al. (1999), in Peptide Nucleic Acids: Protocols and Applications, pages 81-86; Lohse et al. (1997) Bioconjugate Chem. 8, 503). The first method is based on labeling the free (deprotected) PNA after it has been synthesized in solution. In this method, the amino terminus of the PNA is reacted with an activated carboxylic acid or an isothiocyanate. However, additional lysine residues are frequently introduced into the PNA, with these residues then being reacted with fluorescein isothiocyanate (FITC).
In the second method, the protected PNA is modified at its amino terminus with activated carboxylic acid derivatives or isothiocyanates while it is still on the solid phase. This method is only suitable for labeling groups which are stable under the conditions which pertain during deprotection of the PNA and during its cleavage from the support. The reactive reagents which are preferably used in both cases are isothiocyanates (P. Wittung et al., (1995) FEBS Lett. 375, 27) and activated carboxylic acids, such as N-hydroxysuccinimide esters (NHS) (Oerum et al., 1999). A disadvantage of the reaction using the NHS derivatives is that it is frequently only accomplished with poor yields. For this reason, 8-amino-3,6-dioxaoctanoic acid is frequently condensed, as a linker or spacer, between the PNA and the labeling group (Oerum et al., 1999). Both linkages are effected by way of amide bonds or thiourea bonds, which, as such, are, however, more likely to lead to insolubility. Alternatively, the carboxylic acids are caused to react using activators which are customary in peptide chemistry, such as HBTU, TBTU or HATU. 
In a third method, shown generally above, fluorescein-conjugated monomers are used during the synthesis of the PNA on the solid phase, with the fluorescence labeling being effected by way of an amide bond (Lohse et al. (1997) Bioconjugate Chem. 8, 503), which once again leads to conjugates that are relatively difficult to dissolve.
A fourth method uses PNA peptide conjugates in which the peptide moiety forms a substrate for a protein kinase (Koch et al. (1995) Tetrahedron Lett. 36, 6933). In this way, therefore, it is not the PNA moiety which is modified; rather, the serine residue in the peptide segment is phosphorylated enzymatically. When this method is used, therefore, it is only possible to introduce radioactive phosphate, and not, for example, any fluorescein or biotin, into the peptide segment of the PNA-peptide conjugate. The general reaction is depicted as follows: 
It is known that PNA tends to aggregate in aqueous solution, that is, under physiological conditions as well. PNA is therefore poorly soluble in aqueous buffer and is then unavailable for hybridizing to complementary sequences. Furthermore, PNA has a high affinity for various materials such as SEPHADEX® (from Pharmacia), BOND ELUT® (from Varian), or various HPLC chromatograph materials that are used in purifying oligomers. This means that PNA can frequently only be isolated in poor yields. It is therefore necessary to conjugate PNA with lysine or other positively charged amino acids (by way of the C terminus) (Egholm et al (1992) J. Am. Chem. Soc. 114, 1895). Guanine-rich PNA sequences have a very particular tendency to aggregate. For this reason, use of such PNA is generally discouraged (see “Guidelines for sequence design of PNA oligomers” in Peptide Nucleic Acids: Protocols and Applications (1999) pages 253-255). For example, relatively long fluorescein-labeled PNA oligomers are particularly difficult to dissolve, with the addition of an organic solvent and heating to 50° C. being recommended.
It is particularly difficult to purify the poorly soluble lipophilic PNA derivatives. Several peaks due to PNA aggregates are frequently detected in the HPLC. The technique of polyacrylamide (PAA) gel electrophoresis, which is frequently employed for purifying and separating nucleic acids, cannot be used for these PNA derivatives.
In the methods of derivatizing PNA which are described above, the labeling group is always introduced by forming an amide bond or a thioamide bond, with PNA derivatives being formed which are relatively difficult to dissolve. Poorly soluble PNA derivatives are formed, in particular, when lipophilic residues, such as fluorescein, are introduced. Inserting labels at both ends of the PNA is technically even more difficult and generally leads to even poorer solubility. In addition, no efficient method for simultaneously derivatizing PNA at the amino and carboxy termini, in particular by means of solid phase synthesis, has been described. Furthermore, since the labeling reactions frequently proceed with poor yields, there is a need in the art to develop PNA derivatives that can be prepared in high yields, and which should exhibit advantageous properties, such as improved solubility, improved binding behavior, and better cellular uptake, and which, in addition, make it possible to use efficient methods for purifying the PNA oligomers.